Zeroshot Text-to-Speech F5-TTS
Text-to-Speech
PiAPI Text-to-Speech API Documentation#
Overview#
PiAPI's Text-to-Speech (TTS) service provides voice cloning and speech synthesis capabilities using zero-shot learning technology. This service allows you to generate speech audio using a reference voice sample, making it possible to create natural-sounding speech in any voice.Pricing#
| Service | Price (USD) |
|---|
| Text-to-Speech | $0.025 per 1000 characters |
Currently supported model:Capabilities: Voice cloning and text-to-speech synthesis
Request
Body Params application/json
Model name for TTS generation
Task type for TTS generation
Text to be converted to speech
Reference audio URL or base64 encoded audio data
Optional text corresponding to the reference audio
Service mode for the request
{
"model": "Qubico/tts",
"task_type": "zero-shot",
"input": {
"gen_text": "string",
"ref_audio": "http://example.com",
"ref_text": "string"
},
"config": {
"service_mode": "public",
"webhook_config": {
"endpoint": "http://example.com",
"secret": "string"
}
}
}
Request samples
curl --location --request POST 'https://api.piapi.ai/api/v1/task' \
--header 'x-api-key;' \
--header 'Content-Type: application/json' \
--data-raw ''
Responses
Model used for generation
Current status of the task
{
"code": 200,
"message": "string",
"data": {
"task_id": "736fde4d-9029-4915-8189-01353d6982cb",
"model": "string",
"task_type": "string",
"status": "string",
"config": {},
"input": {},
"output": {
"audio_url": "http://example.com",
"status": 0,
"message": "string"
},
"meta": {
"created_at": "2019-08-24T14:15:22Z",
"started_at": "2019-08-24T14:15:22Z",
"ended_at": "2019-08-24T14:15:22Z",
"usage": {
"type": "string",
"consume": 0
}
}
}
}
Modified at 2025-04-03 02:46:55
